Table 3:

Impact of changing the mammogram recommendation age from 40 to 45

	Rec at Age 40	Rec at Age 45	Change
A. Screening and spending (per woman)
Mammograms	4.70 (0.06)	3.80 (0.14)	−0.90 (0.08)
Negative	4.22 (0.05)	3.42 (0.12)	−0.81 (0.07)
False positives	0.46 (0.01)	0.36 (0.02)	−0.09 (0.01)
True positives	0.0208 (0.0024)	0.0204 (0.0024)	−0.0004 (0.0001)
In-situ diagnoses	0.0063 (0.0005)	0.0060 (0.0005)	−0.0004 (0.0001)
Invasive diagnoses	0.0145 (0.0019)	0.0145 (0.0019)	0.0000 (0.0001)
Total health care spending ($)	71,326 (128)	71,007 (155)	−319 (29)
B. Mortality (per 1,000 women by age 50)
Dead	15.98 (0.53)	16.03 (0.53)	0.05 (0.03)
Dead from breast cancer	8.23 (0.53)	8.28 (0.53)	0.05 (0.03)
Dead from other reason	7.75 (0.00)	7.75 (0.00)	0.00 (0.00)
Years alive, per woman	15.87 (0.00)	15.87 (0.00)	−0.0002 (0.0001)

Table reports model predictions for various outcomes under the status quo recommendation that mammograms begin at age 40 (column 1) and the counterfactual recommendation that mammograms begin at age 45 (column 2). The predictions are generated using the parameter estimates from Table 2, and simulated women’s life histories under a non-screening regime based on the clinical oncology model. Panel A reports the average number of mammograms and different mammogram outcomes per woman over ages 35–50. Panel B shows the share of women dead (and from different causes) by age 50, as well as the number of years alive on average between 35 and 50. Standard errors are calculated using 100 repetitions of the bootstrap.